Esteban Montenegro-Montenegro, PhD
Psycholoy and Child Development
| Quantitative | Qualitative | Mixed Methods |
|---|---|---|
| Experimental designs | Narrative Research | Convergent |
| Non-experimental | Phenomenology | Explanatory sequential |
| Longitudinal Designs | Grounded Theory | Exploratory sequential |
| Ethnographies | Complex designs with embedded core designs | |
| Case Study |
Survey research: provides a quantitative or numeric description of trends, attitudes, or opinions of a population by studying a sample of that population. It includes cross-sectional and longitudinal studies using questionnaires or structured interviews for data collection—with the intent of generalizing from a sample to a population.
Experimental research: seeks to determine if a specific treatment influences an outcome. The researcher assesses this by providing a specific treatment to one group and withholding it from another and then determining how both groups scored on an outcome.
Qualitative designs
Narrative research: The information retold or restoried by the researcher into a narrative chronology. Often, in the end, the narrative combines views from the participant’s life with those of the researcher’s life in a collaborative narrative
Phenomenological research: the researcher describes the lived experiences of individuals about a phenomenon as described by participants.
Grounded theory: is a design of inquiry from sociology in which the researcher derives a general, abstract theory of a process, action, or interaction grounded in the views of participants.
Ethnography: is a design of inquiry coming from anthropology and sociology in which the researcher studies the shared patterns of behaviors, language, and actions of an intact cultural group in a natural setting over a prolonged period of time.
Case studies: in-depth analysis of a case, often a program, event, activity, process, or one or more individuals.
Let’s take a look at some spurious correlations:
Can we know if A causes B with a survey?
Can we know if A causes B conducting an experiment?
We can manipulate a variable and observe what happens afterwards, but it is good enough?
Do we need something more on our design?
Shadish et al. (2002) :
Shadish et al. (2002) :
Important
Many factors are usually required for an effect to occur, but we rarely know all of them and how they relate to each other. This is one reason that the causal relationships we discuss are not deterministic but only increase the probability that an effect will occur.
What is an effect?
Important
We could add a group of participants to a waiting list, do you have any example in mind?
Shadish et al. (2002) :
Warning
Correlation does not prove causation!!! We will use this as a mantra in this class.
Don’t forget how variable is the concept of variable!
Moderating variables are predictor variables that affect the direction and/or the strength of the relationship between independent and the dependent variable.
Shadish et al. (2002):
A “pure experiment” is a design where participants are selected randomly, and assign to conditions randomly.
The randomization creates two or more groups of units that are probabilistically similar to each other on the average.
If we are flexible about the randomization, we are designing a study closer to a quasi-experiment.
Quasi-experiments also test descriptive causal hypothesis, there might be pretests, control groups, but they lack random assignment.
The assignment can be done by self-selection, or by administrator selection.
Let’s imagine you need to study the effects of smocking tobacco on your working memory, can you do a random experiment with human subjects? Would it be better to do a quasi-experiment?
Shadish et al. (2002):
| Type of validity | Concept |
|---|---|
| Statistical Conclusion Validity | The validity of inferences about the correlation (covariation) between treatment and outcome. |
| Internal Validity | The validity of inferences about whether observed covariation between A (the presumed treatment) and B (the presumed outcome) reflects a causal relationship from A to B as those variables were manipulated or measured. |
| Construct Validity | The validity of inferences about the higher order constructs that represent sampling particulars. |
| External Validity | The validity of inferences about whether the cause-effect relationship holds over variation in persons, settings, treatment variables, and measurement variables. |